Urban Tree Height Assessment: A Case Study in Washington, D.C.

Abstract

This study evaluates the effectiveness of Light Detection and Ranging (LiDAR) technology for detecting individual urban trees and estimating their heights, focusing on Ward 7 in Washington, D.C. Using LiDAR data collected in 2015 and 2020, we applied tree-finding algorithms to identify individual tree locations and assessed detection performance and height-estimation accuracy against reference measurements. The 2020 dataset detected 47.6% of known trees, compared with 32% for the 2015 dataset, and LiDAR-derived heights showed a strong correlation with reference measurements (R² = 0.79) in the 2020 data. The findings support LiDAR as a viable tool for urban forestry management, including monitoring canopy growth over time and informing planting and maintenance planning.

Background

Urban tree canopy contributes to environmental sustainability, human health, and mitigation of the urban heat island effect. The District of Columbia has stated a goal of achieving 40% tree coverage by 2032, which requires detailed and continuous monitoring of the city’s trees. The District’s urban forestry division maintains a database of nearly 175,000 city-managed trees, but this database primarily records species and location, with limited continuous data on canopy height — a gap that can hinder effective tree management.

LiDAR, a remote sensing method that uses laser pulses to generate precise three-dimensional information about surface features, has become an increasingly viable tool for urban tree canopy assessment. Beyond detecting individual trees, LiDAR can measure canopy height, supporting growth monitoring, planting planning, and impact assessment.

Study Area & Research Questions

The study concentrated on Ward 7, in the easternmost part of the District of Columbia, including the Kingman Park and Hill East neighborhoods. The Ward includes historic parks and the Anacostia River Park and had a population of 89,870 as of the most recent study period, having grown from 71,748 to 76,255 people between 2010 and 2020 (a 6.3% increase). Using LiDAR data collected in 2015 and 2020, the research team sought to answer three questions:

• Can LiDAR effectively detect tree locations in a densely populated urban area?
• Can LiDAR accurately estimate tree heights, and how does it compare to traditional ground-survey measurements?
• Can LiDAR be used to estimate changes in tree height over time?

Key Findings

The tree-finding algorithm effectively detected individual urban trees above 2 meters in height. The 2020 LiDAR data successfully identified 47.6% of known trees in Ward 7, compared with a lower detection rate of 32% using the 2015 data — an improvement consistent with advances in LiDAR data quality and resolution between acquisitions.

LiDAR-derived tree heights showed a strong correlation with reference measurements, particularly in the 2020 data, which produced an R² value of 0.79. The research also identified inconsistencies between LiDAR-detected tree locations and the District’s own tree database, suggesting an opportunity to refine the city’s existing tree-detection methods.

Implications for Urban Forestry Management

The ability to monitor changes in tree height over time using LiDAR is particularly valuable for urban forestry management — enabling city planners to track canopy growth, assess the effectiveness of maintenance programs, and plan future plantings with greater precision. As the District pursues its canopy-coverage goals, accurate, repeatable measurement of tree height and distribution will support broader environmental initiatives, including stormwater management, urban heat island mitigation, and biomass estimation.

Conclusion

This case study demonstrates that LiDAR is a viable, increasingly accurate tool for urban tree canopy assessment at the individual-tree level. As acquisition technology and data resolution continue to improve, LiDAR-based approaches offer cities like Washington, D.C. a scalable path to more precise, data-driven urban forestry management.